The DOI matches its Crossref record; 76 of the 76 references it cites were found in a registry or an index; 2 of the 19 sources its claims cite were checked for support.
Time Delay Estimation in Room Acoustic Environments: An Overview
Open accessAbstract
Time delay estimation has been a research topic of significant practical importance in many fields (radar, sonar, seismology, geophysics, ultrasonics, hands-free communications, etc.). It is a first stage that feeds into subsequent processing blocks for identifying, localizing, and tracking radiating sources. This area has made remarkable advances in the past few decades, and is continuing to progress, with an aim to create processors that are tolerant to both noise and reverberation. This paper presents a systematic overview of the state-of-the-art of time-delay-estimation algorithms ranging from the simple cross-correlation method to the advanced blind channel identification based techniques. We discuss the pros and cons of each individual algorithm, and outline their inherent relationships. We also provide experimental results to illustrate their performance differences in room acoustic environments where reverberation and noise are commonly encountered.
Overview
CoChat Summary
TL;DR
This paper systematically reviews time-delay estimation (TDE/TDOA) algorithms for room acoustic environments, from cross-correlation methods to blind channel identification techniques. It also compares eight algorithms experimentally under simulated noise and reverberation, finding that multichannel and channel-identification methods generally outperform dual-sensor correlation methods in adverse conditions.
Key findings
- The review classifies TDE algorithms by number of sources, propagation model (ideal single-path, multipath, reverberation), analysis tool (e.g., GCC, HOS, blind channel identification), and whether the delay estimate is adaptive or nonadaptive.
- In moderate noise (SNR = 15 dB) and reverberation times of 120 ms, 350 ms, and 580 ms, all tested algorithms accurately identified the delay at 120 and 350 ms; at 580 ms, CC and ML degraded significantly, PHAT remained reasonably robust, AED was best among the five dual-sensor techniques, and AMC achieved the best overall performance.
- In nonreverberant noisy conditions (SNR = 15 dB, 5 dB, and −5 dB), all techniques worked well down to 5 dB; at −5 dB performance deteriorated, with LMS most sensitive to noise, PHAT inferior to CC and ML in additive noise, ML better than CC, AED most robust among dual-sensor methods, and MCCC better than the fusion method.
- In tracking a source moving from (100, 100, 40) to (32, 100, 40), with true delay changing from 5 to 3 samples at SNR = 0 dB and T60 = 240 ms, all algorithms adjusted to the new delay in less than one second.
Methods
- Design: a review/overview of TDE algorithms plus simulation experiments comparing eight algorithms: five dual-channel techniques (CC, PHAT, ML, AED, LMS) and three multiple-channel techniques (MCCC, AMC, FUSION).
- Data/sample: room acoustic simulations using the image model with a rectangular room of 120 × 180 × 150 inch, reflection coefficients varying between 0 and 1, two point omnidirectional sources at (100, 100, 40) and (32, 100, 40), and a four-microphone linear array at (20, 10, 40), (28, 10, 40), (36, 10, 40), and (44, 10, 40); a female speech signal digitized at 16-bit/16 kHz was convolved with synthetic impulse responses, and white Gaussian noise was added to control SNR from −10 dB to 25 dB.
- Main analysis: delay estimates were obtained frame-by-frame with 64 ms frames; cost functions were smoothed by a single-pole recursion; each algorithm was optimized in a nonreverberant, weak-noise condition (SNR = 25 dB) before testing in reverberation and different noise conditions, with results shown as delay histograms and tracking plots.
Limitations
- Stated: the tolerance of TDE to distortion, especially reverberation, remains an open problem; multichannel and fusion methods have higher computational complexity and require synchronization of signals or cost functions; resolution is limited by the sampling period unless interpolation, higher sampling rate, or larger sensor spacing is used, each with tradeoffs.
- Stated: the ML processor’s optimality requires a large observation sample space, no reverberation, constant delay, stationary observation signals, and known noise spectra; otherwise it becomes suboptimal like other GCC members.
- Evident from the text: the experiments are simulation-based using the image model with frequency- and incidence-independent uniform reflection coefficients and ideal point microphones, not real-room recordings; performance comparisons are based on one female speech source and a specific array/room configuration, so generalization to other real acoustic conditions is not established by the reported experiments.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
2 Valid · 5 Not valid · 10 Not checked · 81 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“As a result, the AMDFapproach should exhibit a similar performance to the CC method from a statistical point of view [43].”
What the sources say
- Partly supportedcites [43] Chen J, Benesty J, Huang Y(Arden): Performance of GCC- and AMDF-based time-delay estimation in practical reverberant environments. EURASIP Journal on Applied Signal Processing 2005, 2005(1):25-36. 10.1155/ASP.2005.25 sourceExperimental — accuracy not yet measured
Section 3.2.3 (page unavailable) reports almost identical biases for GCC and AMDF-based estimators, supporting similarity in one statistical measure. But Section 3.2.2 reports different deviations as noise increases, and Section 2.2.2 says AMDF is less robust to noise. The paper therefore does not support a general expectation of similar statistical performance. (LLM, conf=0.82)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“When reverberation time is increased to 580 ms, both the CC and the ML methods suffer significant performance degradation”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 3 assumptions not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Among all the eight algorithms studied, the LMS method is most sensitive to noise.”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 2 assumptions not shown here.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The MCCC shows a better performance than the fusion method.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 2 assumptions not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For each data sample, the number of multiplications required for computing the cost function is shown in Algorithm 2, which is higher than that of the PHAT algorithm.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
47 more claims
Create a free account to see all 98 claims and every source verdict.
- It may state two separate claims as one. 24
- The quote only partly states this claim. 3
- Probably not a substantial claim. 27
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ 6.3. Experimental results · p. 14
“In this case, the true delay in the first 30 seconds is 5 (samples), and 3 (samples) then.”
§ 6.1. Experimental setup · p. 11
“- (iii) Source position: two point omnidirectional sources are located at (100, 100, 40) and (32, 100, 40), respectively. - (iv) Sensor positions: a linear array which consists of four (4) ideal point microphones placed in parallel with the x -axis. The four microphones are located at (20, 10, 40), (28, 10, 40), (36, 10, 40), and (44, 10, 40), respectively.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 7. SUMMARY · p. 14
“In case that the true delay is not the integral multiple of the sampling rate, we will have to either increase the sampling rate or use interpolation, which may significantly increase the computational complexity.”
§ 5. RESOLUTION PROBLEM · p. 9
“The delay estimate is, therefore, an integral multiple of the sampling period.”
Citations
Cites
The generalized correlation method for estimation of time delay
Knapp, C., & Carter, G. (1976) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/tassp.1976.1162830 · Cited by 4,433
Image method for efficiently simulating small-room acoustics
Allen, J. B., & Berkley, D. A. (1979) · The Journal of the Acoustical Society of America · doi:10.1121/1.382599 · Cited by 3,742
Livescu, K., Bezman, A., Borges, N., Yung, L.-Y., Çetin, Ö., Frankel, J., King, S., Magimai.-Doss, M., Chi, X., & Lavoie, L. (2007) · Edinburgh Research Explorer (University of Edinburgh) · Cited by 2,145
Coherence and time delay estimation
Carter, G. C. (1987) · Proceedings of the IEEE · doi:10.1109/proc.1987.13723 · Cited by 1,200
Parameter estimation of superimposed signals using the EM algorithm
Feder, M., & Weinstein, E. (1988) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/29.1552 · Cited by 779
A least-squares approach to blind channel identification
Xu, G., Liu, H., Tong, L., & Kailath, T. (1995) · IEEE Transactions on Signal Processing · doi:10.1109/78.476442 · Cited by 777
Robust Localization in Reverberant Rooms
DiBiase, J. H., Silverman, H. F., & Brandstein, M. S. (2001) · Digital Signal Processing · doi:10.1007/978-3-662-04619-7_8 · Cited by 678
Multichannel blind identification: from subspace to maximum likelihood methods
Tong, L., & Perreau, S. (1998) · Proceedings of the IEEE · doi:10.1109/5.720247 · Cited by 572
Time delay estimation for passive sonar signal processing
Carter, G. (1981) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/tassp.1981.1163560 · Cited by 534
Real-valued fast Fourier transform algorithms
Sorensen, H., Jones, D., Heideman, M., & Burrus, C. (1987) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/tassp.1987.1165220 · Cited by 500
Research works printed but not linked: 1
- Dvorkind TG, Gannot S: Approaches for time different of arrival estimation in a noisy and reververant environment. Proceedings of International Workshop on Acoustic Echo and Noise Control (IWAENC '03), September 2003, Kyoto, Japan 215-218.
Cited by
A Consolidated Perspective on Multimicrophone Speech Enhancement and Source Separation
Gannot, S., Vincent, E., Markovich-Golan, S., & Ozerov, A. (2017) · IEEE/ACM Transactions on Audio, Speech, and Language Processing · doi:10.1109/taslp.2016.2647702 · Cited by 563
Real-Time Multiple Sound Source Localization and Counting Using a Circular Microphone Array
Pavlidi, D., Griffin, A., Puigt, M., & Mouchtaris, A. (2013) · IEEE Transactions on Audio, Speech, and Language Processing · doi:10.1109/tasl.2013.2272524 · Cited by 254
Do, H., Silverman, H. F., & Yu, Y. (2007) · 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07 · doi:10.1109/icassp.2007.366631 · Cited by 223
“The pros and cons of many proposed location algorithms have been discussed in earlier work [1, 2, 3, 4, 5].”
BackgroundJoint TOA and AOA Estimation for UWB Localization Applications
Taponecco, L., D'Amico, A. A., & Mengali, U. (2011) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2011.042211.100966 · Cited by 187
“Before discussing how to improve τΔ it is worth mentioning a method described in [33] to estimate τΔ in room acoustic environments.”
BackgroundMethodInfluentialCobos, M., Marti, A., & Lopez, J. J. (2011) · IEEE Signal Processing Letters · doi:10.1109/lsp.2010.2091502 · Cited by 183
Metadata
- Authors
- Chen, J., Benesty, J., & Huang, Y.
- Affiliations
- Institut National de la Recherche Scientifique (CA)
- Corresponding author
- Chen, Jingdong
- Venue
- EURASIP Journal on Advances in Signal Processing
- Published
- 2006-05-03
- Volume
- 2006
- Issue
- 1
- ISSN
- 1110-8657 (ISSN-L)
- 1687-6180
- OpenAlex
- W2116635162
- Semantic Scholar
- 2edb026228a5050cd52778d04c52d7e7a45324b9
- Cited by
- 370
- References
- 69
- Citations by year
- 2026: 7 · 2025: 10 · 2024: 10 · 2023: 10 · 2022: 11 · 2021: 22 · 2020: 18 · 2019: 27 · 2018: 19 · 2017: 20 · 2016: 27 · 2015: 26 · 2014: 17 · 2013: 28 · 2012: 33
- FWCI
- 21.24
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- Locations
- EURASIP Journal on Advances in Signal Processing (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Physical Sciences › Computer Science › Signal Processing › Speech and Audio Processing
- Physical Sciences › Computer Science › Signal Processing › Direction-of-Arrival Estimation Techniques
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Keywords
- time delay estimation, room acoustics, reverberation, noise robustness, cross correlation function, blind channel estimation, source localization, source tracking, source identification
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Time Delay Estimation in Room Acoustic Environments: An Overview. https://app.cochat.ai/papers/10.1155/asp/2006/26503
- The paper
- Chen, J., Benesty, J., & Huang, Y. (2006). Time Delay Estimation in Room Acoustic Environments: An Overview. EURASIP Journal on Advances in Signal Processing. https://doi.org/10.1155/asp/2006/26503
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.